Multiomic characterization, immunological and prognostic potential of SMAD3 in pan-cancer and validation in LIHC

doi:10.21203/rs.3.rs-3780370/v1

Download PDF

Research Article

Multiomic characterization, immunological and prognostic potential of SMAD3 in pan-cancer and validation in LIHC

https://doi.org/10.21203/rs.3.rs-3780370/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background：SMAD3, a protein-coding gene, assumes a pivotal role within the transforming growth factor-beta (TGF-β) signaling pathway. Notably, aberrant SMAD3 expression has been linked to various malignancies. Nevertheless, an extensive examination of the comprehensive pan-cancer impact on SMAD3's diagnostic, prognostic, and immunological predictive utility has yet to be undertaken.

Methods： Bioinformatics methods were employed to systematically investigate the potential carcinogenic impact of SMAD3. We extensively harnessed data from authoritative sources, including The Cancer Genome Atlas (TCGA), Genotype-Tissue Expression (GTEx), cBioPortal, Human Protein Atlas (HPA), UALCAN, and various other databases. Our study encompassed a comprehensive analysis of the following aspects: differential SMAD3 expression and its association with prognosis across diverse cancer types, gene mutations, immune cell infiltration, single-cell sequencing analysis, DNA methylation patterns, and drug sensitivity profiles. In vitro experiments were conducted with the primary objective of appraising both the expression profile and the precise functional attributes of SMAD3 within the milieu of Liver Hepatocellular Carcinoma (LIHC).

Results：Our findings revealed significant variations in SMAD3 expression between cancerous and adjacent normal tissues. High levels of SMAD3 expression were consistently associated with unfavorable prognoses across multiple cancer types,. Additionally, our analysis of SMAD3 methylation patterns in human cancers unveiled a favorable prognosis linked to elevated DNA methylation levels in pan-cancer. Furthermore, we identified positive associations between SMAD3 expression and RNAm6A methylation-related genes in the majority of cancers. Moreover, SMAD3 expression displayed substantial correlations with immune cell infiltration. Notably, immune checkpoint genes exhibited significant associations with SMAD3 expression across diverse cancers. Single-cell sequencing results elucidated the pan-cancer single-cell expression landscape of SMAD3. Within specific cancer subtypes, SMAD3 expression exhibited a noteworthy positive association with distinctive facets of malignancy. Finally, in our comprehensive analysis of drug sensitivity, we discerned a catalog of prospective therapeutic agents.

Conclusions：In our comprehensive analysis across multiple cancer types, we observed a significant disparity in SMAD3 expression compared to normal tissues, and this These findings suggest that SMAD3 holds promise as both a prognostic biomarker and a therapeutic target against various cancers. Difference displayed a noteworthy association with patient prognosis.

smad3

pan-cancer

TGF-β

LIHC

diagnosis

prognosis

immunization

Cancer ranks as the second leading global cause of mortality, responsible for approximately 14.1 million new cases and 8.2 million annual fatalities (1). While recent years have witnessed the emergence of immunotherapy and targeted therapy as novel cancer treatment strategies, their applicability remains limited to a subset of patients (2). Moreover, the five-year survival rates continue to be relatively modest (3). Consequently, enhancing the effectiveness of cancer treatment necessitates the identification of more potent biomarkers (4).

SMAD3 is a pivotal transcription factor that modulates the transforming growth factor-beta (TGF-β) signal and assumes a dual role in both restraining and promoting cancer. It features highly conserved regions at the N- and C-termini, known as the mad-homology domain 1 (MH1) and MH2 regions, respectively. The MH1 region primarily governs DNA binding, while the phosphorylated MH2 region plays a critical role in the canonical TGF-β signaling pathway, As the C-terminal phosphorylated form, known as R-SMAD, SMAD3 functions as an intracellular signaling transducer and transcriptional regulator. Activation is initiated by TGF-β and activin type 1 receptor kinases. R-SMAD binds to TGF-β responsive elements (TREs) in the promoter regions of numerous TGF-β-regulated genes. Following the formation of the SMAD3/SMAD4 complex, it translocates to the nucleus to activate transcription, modulate gene activity, and stimulate cell proliferation.

Numerous studies have illuminated SMAD3's involvement in various disease states, with a growing recognition of its role in cancer(5). Research has revealed that SMAD3 promotes the formation of cancer-associated fibroblasts through macrophage-myofibroblast transition. Chromatin immunoprecipitation sequencing (ChIP-seq) has demonstrated substantial enrichment of Smad3 binding to genes related to fibroblast differentiation in LLC tumor macrophage lineage cells. Notably, targeted deletion in macrophages and pharmacological inhibition of Smad3 effectively thwart macrophage-myofibroblast transition, thereby impeding cancer-associated fibroblast formation and cancer progression in vivo (6). Aberrant SMAD3 activation is a pivotal driver of breast cancer metastasis. This occurs via the acetylation of SMAD3 by KAT3A at K20 and K117, promoting interactions with TRIM24, and disrupting the interaction with the tumor suppressor TRIM33. Consequently, this leads to heightened SMAD3 activation and the expression of immune response-related cytokines, culminating in increased breast cancer stemness, myeloid-derived suppressor cells (MDSC), and metastasis in triple-negative breast cancer (TNBC) (7). Furthermore, SMAD3 plays a role in promoting cancer progression by impeding E4BP4-mediated natural killer (NK) cell development (8). Subsequent studies unveiled that NK-3-S1KD, a genetically engineered stable Smad3-silenced human NK cell line, effectively curbs cancer progression in xenografted mouse models with human liver cancer and melanoma (9). Elevated SMAD3 expression levels also correlate with the progression and prognosis of various cancers, including bladder urothelial carcinoma (BLCA), colon adenocarcinoma (COAD), prostate adenocarcinoma (PRAD), and cervical carcinoma (CESC)(10–14). Additionally, numerous drugs have been found to regulate tumor proliferation and metastasis through SMAD3, including CDK4 inhibitors and doxorubicin, which induce breast cancer cell apoptosis via SMAD3. Docetaxel inhibits SMAD3 binding to the HIF-1α gene promoter, thereby curtailing the transcriptional activation of HIF-1 and inhibiting cell proliferation (15, 16). In summation, the significance of SMAD3 in the development of multiple cancers is gaining increasing recognition, rendering it a promising candidate as a novel diagnostic and prognostic biomarker and therapeutic target for these malignancies.

In this study, we conducted a comprehensive bioinformatics analysis of SMAD3-related data. Utilizing a range of databases including HPA, cBioPortal, UALCAN, GSCA, GEPIA, CancerSEA, TISMO, TISCH, and others, we assessed the expression levels of SMAD3 across various tissues and its associations with different cancer types. Furthermore, we delved into the prognostic significance, mutation patterns, and single-cell expression profiles of SMAD3. Our analysis extended to exploring the connections between SMAD3 and immune cell infiltration, immune-related genes, immune checkpoints, immunomodulatory genes, methylation patterns, tumor mutation burden (TMB), microsatellite instability (MSI), tumor stemness, and drug sensitivity. Additionally, we conducted functional enrichment analysis of genes related to SMAD3 to unveil the underlying molecular mechanisms contributing to diverse cancers. This study collectively sheds light on the potential of SMAD3 as a prognostic marker in various cancer contexts. Our in-depth in vitro experiments have provided substantial evidence regarding the atypical expression and biological roles of SMAD3 in Liver Hepatocellular Carcinoma (LIHC). The findings firmly establish SMAD3 as an independent and promising prognostic marker, underscoring its potential as a fresh therapeutic target for LIHC.

Gene Expression Analysis of SMAD3 in Pan-Cancer

The TIMER2.0 database (17) stands as an intricate network for the assessment of immune cell infiltration within tumor tissues. It not only facilitates the scrutiny of immune cell infiltration levels in tumors but also enables the comparative examination of SMAD3 expression profiles across various cancer types and their adjacent normal tissues.

The unified and standardized pan-cancer dataset was acquired through retrieval from the UCSC database (18).Subsequently, the expression data for the SMAD3 gene was meticulously extracted from each sample, culminating in the acquisition of comprehensive expression data spanning 30 distinct cancer species.

The Human Protein Atlas (HPA) constitutes a repository of human proteins within the framework of diverse anatomical compartments, encompassing organs, tissues, and cellular constituents, leveraging a multitude of omics methodologies. The expression of SMAD3 mRNA in normal tissues was detected using the HPA database. In addition, histological and pathological immunohistochemical images of the SMAD3 protein were obtained from the database, and the subcellular localization of SMAD3 was determined through the utilization of indirect immunofluorescence microscopy (19, 20).

Prognostic Analysis of SMAD3

Our data acquisition process entailed the retrieval of a meticulously standardized pan-cancer dataset originating from the UCSC database, specifically the TCGA TARGET GTEx dataset, which comprised a substantial cohort (PAN-CANCER N = 19131 G = 60499). Within this dataset, we meticulously extracted the gene expression data for SMAD3 from each sample. In addition, we secured a high-quality TCGA prognostic dataset, sourced from prior TCGA prognostic studies as published in Cell (21). To complement this dataset, we incorporated the follow-up data obtained from TARGET through UCSC.our study encompassed the examination of various endpoints, including overall survival (OS), disease-specific survival (DSS), disease-free interval (DFI), and progression-free interval (PFI), to scrutinize the association between SMAD3 expression and cancer prognosis. For this purpose, we harnessed the robust analytical tools of Kaplan-Meier survival analysis and Cox regression analysis to delve into the intricate relationship between SMAD3 and survival prognosis (22) .

DNA methylation and RNA modification analysis

The occurrence and proliferation of cancer are regulated by epigenetic inheritance and genetic events, in which DNA methylation plays an important role (23). We used data from the UALCAN database to determine SMAD3 promoter DNA methylation levels to determine the difference between the cancers and normal tissues. We also obtained DNA methylation prognostic data from the GSCA database.

More and more evidence show that RNA methylation is closely related to cancer cell proliferation, cell stress, metastasis, and the immune response. RNA methylation-related proteins have become promising targets for cancer therapy (24). Pan-cancer data were retrieved from the UCSC database, and from this dataset, we extracted the expression profiles of the SMAD3 gene along with 44 marker genes related to Class III RNA modification (comprising m1A (10), m5C (13), and m6A (21) genes) within each sample. Subsequently, Pearson correlation coefficients were computed to elucidate the associations between SMAD3 and these marker genes within the context of five distinct immune pathways.

Gene Mutation Analysis of SMAD3

The cBioPortal database (25) serves as a repository for the comprehensive compilation of data pertaining to mutation frequency, mutation types, specific mutation locations, and the three-dimensional (3D) structural attributes of prospective proteins across the entirety of TCGA cancer datasets.

We downloaded the uniformly standardized pan-cancer dataset from the UCSC database. In addition, we also downloaded the copy number variation dataset at the level-4 gene level for all TCGA samples by GISTIC (26) software from the GDC. We integrated the copy number data and gene expression data of the samples and analyzed the CNV mutation differences between pan-cancers by using R software.

Immune-Related Characteristics of SMAD3

We conducted an extensive analysis of immune cell infiltration by leveraging the resources available in the TIMER 2.0 database. Leveraging a range of algorithms, including TIMER, XCELL, EPIC, CIBERSORT, TIDE, QUANTISEQ, and others, we meticulously examined the associations between SMAD3 expression and diverse immune cell subsets. These subsets encompassed CD8 + T cells, CD4 + T cells, B cells, macrophages, and T cell regulators (Tregs) within distinct TCGA cancer types.

We extracted the expression data of SMAD3 in each sample from the TCGA TARGET GTEx database, and calculated stromal, immune, and estimate scores according to gene expression by using the R software package ESTIMATE(27). In addition, expression data of SMAD3 gene and 150 marker genes of five types of immune pathways (chemokine (41), receptor (18), MHC (21), immunoinhibitor (24) and immunostimulator (46)) were extracted from the database, Correlative studies on SMAD3 expression with immunoregulatory genes were also discussed by Spearman analysis.

In our study, we harnessed Pearson correlation analysis to assess the associations between SMAD3 gene expression and key parameters, including immune checkpoints (ICP), tumor mutation burden (TMB), microsatellite instability (MSI), and tumor stemness. Additionally, we conducted a comparative analysis of SMAD3 expression levels in different tumor cell lines before and after immune checkpoint blockade (ICB) treatment, as well as between responders and non-responders. These analyses were carried out using data from the TIDE database.Furthermore, we examined the correlation between SMAD3 expression and overall survival (OS) within the human immunotherapy cohort and evaluated response outcomes utilizing the TISMO database(28).

Single-Cell Functional Analysis of SMAD3

TISCH, an extensive and meticulously curated database, amalgamates the single-cell transcriptome profiles of approximately 2 million cells sourced from high-caliber tumor datasets spanning 27 distinct cancer types (29). We acquired the single-cell expression data of SMAD3 encompassing various pan-cancer datasets from the database. Our analysis delved into the single-cell composition of SMAD3 expression, and we meticulously examined this within the context of three distinct datasets.

CancerSEA is a database that studies tumor cell function at the single-cell level (30). In our research, we mapped the correlations between SMAD3 expression and diverse tumor functions by leveraging single-cell sequencing data. Furthermore, we provided a demonstration of the spatial transcriptomic characteristics specific to Breast Cancer (BRCA) and Prostate Adenocarcinoma (PRAD) based on data sourced from the SpatialDB database (29).

Co-Expressed Genes and Enrichment Analysis of SMAD3

Our research methodology involved several key steps. Initially, we procured a set of 50 Smad3-interacting genes through the STRING website. Subsequently, to acquire a practical list of Smad3-binding proteins, we utilized the "similar gene detection" feature of GEPIA2. This process relied on the TCGA tumor and normal tissue dataset, from which we extracted the first 100 SMAD3-related genes. Concurrently, we conducted a Pearson correlation analysis to further refine the selection of genes that exhibited a significant correlation with SMAD3. Moving forward, we conducted comprehensive Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses to unravel the intricate functional pathways and biological processes at play. This investigation was undertaken with the aim of elucidating the multifaceted influence of SMAD3 on immune responses and its intricate involvement in the pathogenesis of diverse maladies. Additionally, we explored the enrichment of transcription factors among the associated genes using the Metascape database. Finally, we employed the Gene_Corr module of TIMER2 to generate heatmap data, focusing on the top ten genes associated with SMAD3 across pan-cancer datasets (31), p-value < 0.05 was considered to be statistically significant.

GSEA Enrichment Analysis and Drug Sensitivity Prediction of SMAD3

Within the PAAD and ACC groups, we harnessed Gene Set Enrichment Analysis (GSEA) to probe the biological signaling pathways. In this analysis, we focused on the top 5 terms derived from both the KEGG and HALLMARK databases. Notably, we considered KEGG pathways with notable enrichment results, as indicated by the enrichment score (ES) and gene ratio. Specifically, gene sets with |NES| ≥ 1 and NOM p ≤ 0.05 were deemed to be significantly enriched (32).

In our quest to explore the nexus between drug sensitivity and the expression of SMAD3, we engaged in a rigorous drug sensitivity analysis employing the GSCALite platform. For this endeavor, Spearman correlation analysis served as our indispensable tool. Within our analytical framework, a positive correlation unveiled that neoplastic cell characterized by heightened SMAD3 expression displayed an elevated proclivity towards drug resistance, whereas a negative correlation signified that those same cells manifested an increased receptivity to pharmacological agents.

Cell culture and transfection

The human Liver Hepatocellular Carcinoma (LIHC) cell lines JHH-5, HepG2, SK-HEP-1, Hep3B, Hu-7, LI-7, and LM3 were procured from the Shanghai Institute of Cell Biology, China. The immortalized liver cell line HL-7702 was obtained from Shanghai Fuxiang Biotechnology Co., Ltd. These cell lines were cultivated in DMEM (Gibco) supplemented with 10% FBS (Hyclone) and exposed to 5% carbon dioxide environment at 37°C. We acquired SMAD3 small interfering RNA (siRNA) from GenePharma in Shanghai, China. In LM3 cells, the multiplicity of infection (MOI) was set at 10. To enhance transfection efficiency, Polybrene was employed, and positive cells were subsequently selected using puromycin screening.

Western blot analysis

The Ethics Committee of the Second Affiliated Hospital of Nanchang University granted authorization for the utilization of human tissue in this research. To extract cellular and human tissue lysates, a radioimmunoprecipitation assay buffer (Solarbio, China), fortified with a cocktail of protease inhibitors, was employed. Subsequently, a 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis technique was implemented to facilitate the separation of the resulting pyrolysis products. These products were subsequently transferred onto polyvinylidene fluoride (PVDF) membranes, followed by incubation with primary antibodies. Notably, the primary antibodies employed included SMAD3 (1:1000, 66516-1-Ig, Proteintech, China) and GAPDH (1:2000, 60004-1-Ig, Proteintech, China). In the subsequent steps, PVDF membranes were subjected to further incubation with relevant secondary antibodies. Ultimately, the membranes were exposed to an enhanced chemiluminescence (ECL) substrate (Proteintech, China), and the discernible protein bands were documented via the SH-Cute 523 imaging system for analysis.

Wound Healing and Transwell Experiments

Wound Healing Assay: Hu7, LM3, and Hep3B cells, at a cell density of 2.5×105 per well, were seeded into 6-well plates. Subsequently, transfection of siSMAD3#1 and siSMAD3#2 was conducted in these plates. After a 24-hour incubation period, a 200 mL pipetting head was employed to create cell scratches within the plates. The cells were then provided with serum-free medium. Subsequent images were captured at intervals of 0 hours, 12 hours, and 24 hours using an inverted microscope (IX81, Olympus Company, Japan).

Transwell Assay: Cells were suspended and enumerated using serum-free medium. These cells were then introduced into the upper chamber at a density of 5×104, while the lower chamber received 500 mL of full medium. Following a 24-hour incubation period, the cells were fixed with 4% paraformaldehyde for 10 minutes. Subsequently, 1% crystal violet was utilized to stain the cells for a period of 10 seconds. Any remaining cells in the upper chamber were gently removed, and images were captured using a brightfield microscope (BX53, Olympus Company, Japan). Furthermore, the number of cells that had traversed the chamber was counted across four randomly selected fields under the microscope.

EdU assay

The transfected Hu-7, LM3, and Hep3B cells, with a seeding density of 2 × 104, were plated in 24-well plates and subsequently incubated for a duration of 72 hours. Following this incubation period, the cells were subjected to culture with EdU reagent for a duration of 2 hours. Following these steps, cellular specimens were meticulously subjected to fixation employing a solution of 4% paraformaldehyde and 0.5% Triton X-100. Subsequent to the fixation process, the specimens underwent staining procedures utilizing Hoechst stain. Quantitative assessment of the EdU inclusion rate was accomplished through the utilization of ImageJ software.

Statistical Analysis

Through the online databases mentioned above, the statistical analysis was automatically computed in this study. These results were considered as statistically significant at *P < 0.0001.

Analysis of Gene Expression

Furthermore, we performed an analysis of SMAD3's differential expression in both tumor and normal tissues by leveraging the TIMER 2.0 database. Our investigation revealed a substantial upregulation of SMAD3 expression in six cancer types, including ESCA, STAD, LUSC, LIHC, CHOL, and HNSC. Conversely, we observed significant downregulation in seven other cancers, specifically COAD, BRCA, PRAD, UCEC, KIRC, THCA, and SKCM (Fig. 2B). Simultaneously, we extracted SMAD3 expression data from normal tissues in the GTEx database and integrated it with the data from TCGA. Our analysis unveiled a remarkable up-regulation of SMAD3 in 15 cancer types, which include ACC, CHOL, COAD, ESCA, GBM, HNSC, KICH, LAML, LGG, LIHC, LUAD, LUSC, PAAD, STAD, and TGCT. In contrast, we observed significant down-regulation in 7 other cancer types, specifically BRCA, KIRC, OV, PRAD, SKCM, UCEC, and UCS (Fig. 2C). Additionally, we conducted a paired difference analysis for SMAD3 expression across pan-cancer datasets. Our analysis revealed a significant increase in SMAD3 expression in LIHC, LUSC, OSCC, CHOL, STAD, and HNSC. Conversely, a decrease in SMAD3 expression was observed in BRCA, UCEC, PRAD, and COAD (Fig. 2D).

We employed R software to compute the variations in gene expression across tumor samples at different clinical stages. Utilizing an unpaired Student's t-Test, we performed a significance analysis of the differences between pairs of samples. In addition, we conducted variance analysis to test the differences among multiple groups of samples. Our observations indicated significant differences in eight types of tumors, including LUAD, BRCA, KIPAN, KIRC-PAAD, OV, TGCT, and BLCA (Fig. 2E-F).

Furthermore, we acquired the IHC results from the HPA database to corroborate the protein expression of SMAD3. As depicted in Fig. 2G-H, tumor tissues displayed substantial SMAD3 IHC staining, whereas normal tissues from the liver, lung, pancreas, and brain exhibited low to moderate staining. Notably, these findings align with the respective levels of gene expression. To gain insight into SMAD3's subcellular localization, we conducted immunofluorescence localization in A-549 and U251-MG cells. Our observations indicated that SMAD3 predominantly localized within the nucleus and cytosols.

Prognostic Value of SMAD3

We employed Overall Survival (OS), Disease-Specific Survival (DSS), and Progression-Free Interval (PFI) to investigate the prognostic implications of SMAD3 across various cancer types. Notably, high expression of SMAD3 was associated with unfavorable outcomes in LAML, PAAD, ACC, UVM, and LUAD (Fig. 3A-B). Conversely, SMAD3 exhibited high expression in GBMLGG, KIRC, and KIPAN, which correlated with favorable OS. In terms of DSS, SMAD3 was highly expressed in ACC, PCPG, LUAD, and PAAD, and this high expression was associated with shorter DSS (Supplementary Fig. 1S A-B). Additionally, our results revealed that SMAD3 was overexpressed in ACC, PAAD, STAD, UVM, LIHC, and STES, and this overexpression was linked to a shorter PFI (Supplementary Fig. 2S A-B). Consequently, SMAD3 overexpression was particularly prominent in PAAD and ACC, where OS, DSS, and PFI were notably abbreviated.

Correlation of SMAD3 Expression with DNA Methylation AND RNA modification

The UALCAN online tool and GSCA database furnished a valuable platform for our exploration of SMAD3 promoter methylation levels across different patient groups and normal populations in various cancer types. Notably, our findings indicated a considerable increase in the promoter methylation levels of SMAD3 in eight tumor groups (BRCA, READ, HNSC, KIRP, UCEC, PRAD, COAD, KIRC) compared to their respective normal counterparts (Fig. 4B). Conversely, in seven tumor groups (LUAD, LUSC, TGCT, CHOL, THCA, BLCA, LIHC), the SMAD3 methylation levels were significantly lower than those observed in normal groups. Moreover, we evaluated the impact of SMAD3 methylation expression on the survival prognosis using the GSCA database. The results were rather promising, demonstrating that high SMAD3 methylation expression tended to predict a more favorable prognosis in nearly all statistically significant prognostic analyses (Fig. 4C, Supplementary Fig. 3SD-G). This suggests that SMAD3 may influence the prognosis of cancer patients through its methylation status. Furthermore, we observed a positive correlation between SMAD3 expression and m6A, m5C, and m1A regulatory genes in most cancers, with the exception of NB and CHOL (Fig. 4A). This intriguing finding implies that SMAD3 may play a role in carcinogenesis by affecting the expression of RNA methylation regulatory genes.

SMAD3 mutation in pan-cancer

Our analysis of genetic alterations across various cancers, conducted through the cBioPortal platform, allowed us to delve into the landscape of SMAD3 gene variations in pan-cancer. The results unveiled distinct patterns in different malignancies, with COAD exhibiting the highest frequency of SMAD3 variations at 6.40%, primarily in the form of mutations. Following COAD, the top five cancers with the highest variation frequencies were UCEC (5.48%), SKCM (3.83%), MESO (3.45%), and KICH (3.08%). Notably, MESO was unique in its pattern of variation, characterized by amplifications (Fig. 5A). In further exploration of these variations, we identified missense and truncating mutations as the primary mutation types affecting SMAD3, as depicted in Fig. 5B. These insights into the genetic alterations in SMAD3 shed light on its role in the context of different cancers.

Five independent studies conducted on endometrial, lung, gastric, and colorectal adenocarcinomas revealed that R268C/H missense mutations underwent changes within the MH2 domain. Additionally, SMAD3 mutations were correlated with various post-translational modifications, including phosphorylation, acetylation, ubiquitination, and malonylation. The alterations in the conformation of the SMAD3 protein resulting from the R268C/H mutations are visually depicted in Fig. 5C.

Changes in copy number in parts of the genome are known to be a feature of many cancers(33). We utilized the R software to compute disparities in tumor gene expression across clinical stage-specific samples. Significance analysis for pairwise disparities was conducted using unpaired Wilcoxon rank sum and signed rank tests, while the Kruskal test was employed to assess differences among multiple sample groups. Figure 5D illustrates notable disparities in copy number variations (CNVs) across 14 distinct cancer types (GBMLGG, LGG, CESC, COAD, COADREAD, BRCA, STES, SARC, STAD, HNSC, LIHC, READ, SKCM, UCS).

Analysis of Immune Cell Infiltration and Immunoinfiltration of SMAD3

The UCSC database was used to download a unified and standardized pan-cancer dataset, and the R software package ESTIMATE was used to calculate stromal, immune, and ESTIMATE scores for each patient in each tumor based on gene expression. To ascertain the statistically significant associations pertaining to immune infiltration scores, we computed the Spearman's correlation coefficient between gene expression and immunoinfiltration scores for each tumor sample. This analysis was conducted by means of the corr.test function within the R package. The results showed that LGG, COAD, COADREAD, PRAD, NB, READ, and UVM showed a significant positive correlation among the three immune infiltration scores. At the same time, STES, KIRP, KIPAN, HNSC, KIRC, LUSC, LIHC, WT, BLCA, THCA, and ACC were all negatively correlated (Supplementary Fig. 4S、Figure 5S、Figure 6S).

In recent investigations, a compelling body of evidence has emerged, affirming the intricate relationship between immune cell infiltration and the genesis, progression, and metastatic cascade of an array of malignancies (34, 35). In the quest to discern the interplay of SMAD3 expression with the presence of tumor-infiltrating immune cell populations, we conducted a comprehensive analysis utilizing the TIMER 2.0 database. Employing a repertoire of algorithms, such as TIMER, EPIC, QUANTISEQ, XCELL, CIBERSORT, TIDE, and others, we unveiled noteworthy findings. Our scrutiny unveiled a substantial positive correlation between SMAD3 expression and the infiltration levels of diverse immune cell types, including cancer-associated fibroblasts (CAF), NK cells, neutrophils, monocytes, and endothelial cells (Endo). Additionally, our investigation exposed a robust positive relationship between SMAD3 expression and the infiltration of CD4 + T cells in prostate adenocarcinoma (PRAD), liver hepatocellular carcinoma (LIHC), cervical squamous cell carcinoma (CESC), and diffuse large B-cell lymphoma (DLBC). In contrast, we identified a substantial negative correlation between SMAD3 expression and CD8 + T cell infiltration in glioblastoma (GBM), head and neck squamous cell carcinoma (HNSC), kidney renal papillary cell carcinoma (KIRP), low-grade glioma (LGG), and lung squamous cell carcinoma (LUSC). Additionally, B cell infiltration displayed a noteworthy positive correlation with various cancer types, including breast invasive carcinoma (BRCA), colon adenocarcinoma (COAD), DLBC, GBM, kidney chromophobe (KICH), LIHC, PRAD, skin cutaneous melanoma (SKCM), and testicular germ cell tumors (TGCT). Notably, our analysis indicated no significant immune-related relationship between SMAD3 expression and macrophages, progenitor cells, eosinophils, hematopoietic stem cells (HSC), myeloid-derived suppressor cells (MDSC), follicular helper T cells (FhT), dermal epithelial cells, and mast cells, as illustrated in Fig. 6.

Relationship between SMAD3 and immune checkpoints, immune modulators

In recent years, immune checkpoint inhibitors (ICIs) have achieved significant milestones in the realm of cancer immunotherapy. However, a substantial number of patients continue to exhibit either a lack of response or develop resistance to immune checkpoint inhibitors (36). We observed that SMAD3 displayed associations with multiple immune checkpoint inhibitors and immune checkpoint stimulators. Subsequently, we conducted an investigation into the relationships between SMAD3 and immune checkpoints as well as immune modulators. Across the majority of cancer types, SMAD3 expression displayed a substantial association with the expression profiles of established immune checkpoint molecules, a trend that was consistently observed. Notable exceptions to this pattern included kidney chromophobe (KICH), esophageal carcinoma (ESCA), acute lymphoblastic leukemia (ALL), cholangiocarcinoma (CHOL), and mesothelioma (MESO), where the correlation between SMAD3 expression and immune checkpoint markers was not statistically significant. The encompassed immune checkpoints encompassed lymphocyte-activation gene 3 (LAG3), CD40, cytotoxic T-lymphocyte-associated protein 4 (CTLA4), CD80, programmed cell death 1 (PDCD1), T cell immunoreceptor with Ig and ITIM domains (TIGIT), CD86, and tumor necrosis factor receptor superfamily, member 9 (TNFRSF9) (Fig. 7A). These findings portend a prospect of synergy between SMAD3 and established immune checkpoint molecules. To appraise the influence of SMAD3 expression on immunotherapeutic outcomes, we conducted a comparative analysis of SMAD3 expression levels in distinct tumor cell lines, both antecedent to and post-administration of immune checkpoint blockade (ICB) interventions, and across respondent and non-respondent subpopulations. Eminent among our observations is the salient decrease in SMAD3 expression within CT26 and T11 cells subsequent to anti-CTLA4 and anti-PD-1 treatments. Furthermore, a conspicuous reduction in SMAD3 expression was discerned in CT26 cells displaying favorable responses to anti-PDL-1 treatment, alongside the corresponding downregulation observed in YTN16 cells that manifested a positive response to anti-CTLA4 intervention (Fig. 7C). Subsequently, we proceeded to assess the relationship between SMAD3 expression and the resultant response patterns, as well as overall survival (OS), within the human immunotherapy cohort (Fig. 7B). The findings unveiled that, within the ambit of 24 immunotherapy cohorts, SMAD3 emerged with an Area Under the Curve (AUC) value exceeding 0.5 in nine instances, denoting its predictive utility. Remarkably, SMAD3 exhibited superior predictive efficacy when contrasted with Tumor Mutation Burden (TMB), T cell clonality (T clonality), and B cell clonality (B clonality). Nevertheless, SMAD3's predictive performance, while notable, was surpassed by the predictive value of the CD274 score (AUC > 0.5 in 21 immunotherapy cohorts), the CD8 score (AUC > 0.5 in 18 immunotherapy cohorts), the Interferon Gamma (IFNG) score (AUC > 0.5 in 17 immunotherapy cohorts), and the Merck 18 score (AUC > 0.5 in 16 immunotherapy cohorts). In the aggregate, SMAD3 expression emerges as a potent biomarker, signifying its efficacy in the anticipation of immunotherapeutic responses. Finally, an examination of SMAD3 expression levels across various datasets revealed its pronounced upregulation in METABRIC, ICB-Lauss2017-ACT, Patel 2017, and MDSC, thereby emphasizing its salient role in these contexts (Fig. 7D).

We additionally conducted an analysis of the correlation between SMAD3 and immune regulatory genes across 40 tumor types. This analysis encompassed Major Histocompatibility Complex (MHC), immunostimulator genes, immunoinhibitor genes, chemokines, and chemokine receptors, aimed at gaining deeper insights into the interplay between SMAD3 and immune regulatory genes within the context of cancer. The results revealed a robust association between SMAD3 expression and immune regulatory genes in malignant tumors (Supplementary Fig. 7S).

Relationship between SMAD3 expression and tumor stemness, and genomic heterogeneity

In addition to metastasis and heterogeneity, it is well-established that immune checkpoint gene expression and immune system cell infiltration are intricately linked to tumor stemness (38). Our analysis of SMAD3 expression alongside tumor stemness markers, DNAss and RNAss, revealed noteworthy findings. We observed significant negative correlations between SMAD3 expression and DNAss in four tumor types, specifically COAD, COADREAD, BRCA, and UCEC. Conversely, significant positive correlations between SMAD3 expression and DNAss were identified in several tumor types, including CESC, STES, THYM, THCA, UVM, and HNSC, as depicted in Fig. 7G. Furthermore, our analysis unveiled significant negative correlations between SMAD3 expression and RNAss in 18 tumor types, encompassing LGG, COADREAD, LAML, BRCA, KIRP, STAD, PRAD, UCEC, THYM, LIHC, THCA, READ, OV, TGCT, UCS, BLCA, and CHOL, as illustrated in Fig. 7H.

Cancer stem cells (CSCs) are a population of self-renewing cells that play a pivotal role in tumor development, facilitate metastasis, and contribute to increased resistance to cancer treatments (37). It is important to note that the impact of immunotherapy can be influenced by tumor heterogeneity. Therefore, we conducted an examination of the Pearson's correlation coefficients between SMAD3 and tumor heterogeneity indicators, including Tumor Mutational Burden (TMB) and Microsatellite Instability (MSI). The analysis revealed that SMAD3 expression was significantly negatively correlated with TMB in tumor types such as GBMLGG, BRCA, LUAD, KIPAN, PRAD, KIRC, THCA, SKCM, BLCA, and DLBC. Conversely, SMAD3 expression exhibited a significant positive association with TMB in SARC, HNSC, ACC, THYM, and PAAD (Fig. 7E). Furthermore, the Microsatellite Instability (MSI) displayed a positive correlation with SMAD3 expression in tumor types including GBMLGG, COAD, SARC, KIPAN, LUSC, and UVM. Conversely, a negative correlation was observed between MSI and SMAD3 expression in BRCA, PRAD, HNSC, THCA, and DLBC (Fig. 7F).

Single-cell Analysis of SMAD3

Utilizing the TISCH database, we conducted an in-depth exploration of SMAD3's single-cell expression patterns across diverse cancer types. Our analysis unveiled that SMAD3 exhibited notably high expression levels in various cell types, including CD4Tconv, Treg, Tprolif, CD8T + cells, CD8Tex, B cells, dendritic cells (DC), endothelial cells, and fibroblasts. Particularly striking was the high expression of SMAD3 in monocytes (Mono) and macrophages (Macaro) (Fig. 8A). To delve deeper, we acquired precise SMAD3 expression data within specific cell groups from breast cancer (BRCA), prostate adenocarcinoma (PRAD), and pancreatic adenocarcinoma (PAAD). This information was sourced from datasets GSE161529, GSE143791, and GSE111672. Our analysis revealed the expression of SMAD3 within key cell groups, including CD8 + T cells, monocytes/macrophages, and endothelial cells (Fig. 8B). Our exploration extended to the investigation of SMAD3's functions at the single-cell level using the CancerSEA database. The heatmap results unveiled significant insights into the correlations between SMAD3 and various biological processes. SMAD3 exhibited positive correlations with critical processes such as angiogenesis, hypoxia, inflammation, metastasis, and stemness across diverse cancer types. In contrast, SMAD3 displayed negative correlations with processes related to DNA damage, DNA repair, invasion, and quiescence (Fig. 8C). In addition, our investigation extended to the SpatialDB database, revealing that SMAD3 is consistently expressed in a majority of monocyte- and macrophage-rich tumors. Our analysis determined that SMAD3, in conjunction with monocyte/macrophage markers such as CD68 and CD14, exhibited varying levels of spatial transcriptomic expression patterns in breast cancer (BRCA) and prostate adenocarcinoma (PRAD) (Fig. 8D).

Enrichment of SMAD3-Related Partners

A total of 50 SMAD3-binding proteins were identified, each supported by experimental data acquired through the STRING tool, which illustrates the interaction network of these proteins (Fig. 9A). Utilizing the GEPIA2 tool to integrate TCGA pan-cancer expression data, we identified the top 100 genes that exhibit a correlation with SMAD3 expression. Among these correlated genes, SMAD3 expression displayed positive associations with the top 10 genes, namely, BICD2 (correlation coefficient, R = 0.5), FAM160A1 (R = 0.52), KLF3 (R = 0.49), MAPK6 (R = 0.48), SEMA4B (R = 0.52), SLK (R = 0.48), TMOD3 (R = 0.52), TP63 (R = 0.53), ZDHHC5 (R = 0.49), and PCYT1A (R = 0.48), with all correlations being statistically significant (P < 0.001) (Fig. 9H-I). In an intersection analysis, two of the aforementioned genes, SMURF1 and SP1, were identified as belonging to both the SMAD3-binding proteins and the top correlated genes with SMAD3 expression (Fig. 9B).

To perform KEGG and GO enrichment analyses, the two datasets were combined. KEGG data suggest that the effect of SMAD3 on tumor pathogenesis, which was depicted in Fig. 7C, may involve the "TGF-beta signaling pathway," "Transcriptional misregulation in cancer," and "Human T-cell leukemia virus 1 infection." (Fig. 9C). Additionally, we performed GO enrichment analysis and discovered that SMAD3-interacting genes are related to the biological processes "reaction to endogenous stimuli," "positive regulation of nucleic acid-templated transcription," and "positive regulation "of biological process (BP) (Fig. 9D). Second, we found that Smad3-related genes were primarily enriched in "nuclear part," "nuclear lumen," and "nucleoplasm" in the cellular component (CC) (Fig. 9E), and that "transcription regulatory region DNA binding", "regulatory region nucleic acid binding" and "double-stranded DNA binding" were enriched in cellular function (MF) (Fig. 9F).

Additionally, we conducted an assessment of Smad3-related genes using the Metascape database and examined the results of transcription factor enrichment analysis. The top three enriched transcription factors identified were as follows: "GGGYGTGNY UNKNOWN", "PAX Q6" and "CEBPB 01" (Fig. 9G).

GSEA Enrichment of SMAD3 and Relationship between SMAD3 expression and drug sensitivity

We performed functional enrichment analysis of SMAD3 expression in pancreatic adenocarcinoma (PAAD) and adrenocortical carcinoma (ACC) using Gene Set Enrichment Analysis (GSEA). In PAAD, SMAD3 expression was predominantly associated with KEGG enrichment terms related to Glycine, Serine, and Threonine Metabolism, the Renin-Angiotensin System, Apoptosis, and Neuroactive Ligand Receptor Interaction (Fig. 10A). Additionally, Apoptosis, Notch Signaling Pathway, TNFA Signaling VIA NFKB, P53 Pathway, and Interferon Gamma Response in PAAD were all linked to SMAD3 in HALLMARK terms (Fig. 10C). According to KEGG enrichment terms, Wnt Signaling Pathway and Notch Signaling Pathway are primarily associated with SMAD3 expression in ACC (Fig. 10B). In ACC, the analysis of SMAD3 expression revealed significant associations with HALLMARK enrichment terms. Specifically, SMAD3 was linked to processes such as Mitotic Spindle, Wnt-β Catenin Signaling, DNA Repair, TGF-β Signaling, and G2M Checkpoint (Fig. 10D).

We assessed the drug sensitivity associated with SMAD3 expression in various cancers using GSCALite. The analysis revealed that the 50 percent inhibitory concentration (IC50) values of nine drugs, namely, 17-AAG, Dasatinib, Docetaxel, Lapatinib, PD-0325901, RDEA119, Trametinib, Selumetinib, and Bleomycin, exhibited an inverse correlation with KIF18A expression. On the other hand, the IC50 values of 15 different medications, including AR-42, BX-912, CAY10603, FX866, GSK1070916, I-BET-762, KIN001-102, NPK76-II-72-1, Navitoclax, QL-XL138, THZ-2-102-1, UNC0638, Vorinostat, WZ3105, and YM201636, were positively correlated with SMAD3 expression (Fig. 10E).

SMAD3 promotes the proliferation and Migration of liver cancer cells

The protein expression levels of SMAD3 in Liver Hepatocellular Carcinoma (LIHC) tissues exhibited a notable elevation, with the outcomes meticulously analyzed utilizing ImageJ software (Fig. 11A). We extended our scrutiny to assess the protein expressions of SMAD3 in a spectrum of LIHC cell lines, encompassing JHH-5, HepG2, SK-HEP-1, Hep3B, Hu7, Li-7, and LM3. Strikingly, we observed that the expression of SMAD3 in the Hep3B, Hu7, and LM3 cell lines surpassed that in other LIHC cell lines (Fig. 11B).

Subsequently, we employed specific siRNAs targeting SMAD3 to achieve knockdown of SMAD3 expression in the Hep3B, Hu7, and LM3 cells (Fig. 11C, D, E). Following this, In our pursuit of unraveling the intricate interplay between LIHC cells and SMAD3 expression across discrete cohorts, we embarked on a series of functional experiments. Notably, the outcomes derived from EDU assays, wound healing assays, and Transwell assays synergistically unveiled a compelling narrative. The act of SMAD3 knock-down ushered forth a marked diminution in the proliferation and migratory potential of Hep3B, Hu7, and LM3 cells (Fig. 11F、G、H). These findings compellingly underscore the promoting role of SMAD3 in the proliferation and migration of LIHC cells.

The similarities and differences between tumors can be shown by pan-cancer analysis, which can support to the theoretical framework guiding efforts in cancer prevention, the design of therapeutic targets, and the prospective screening of therapeutic drugs (38). SMAD3 is central to the TGF-β pathway and plays an important role in immune-inflammatory responses. Recent research suggests that SMAD3 is connected to the survival and prognosis of malignant tumors and plays a role in the etiology and advancement of colorectal cancer, lung cancer, breast cancer, bladder cancer, prostate cancer, cervical cancer, gastric cancer, and osteosarcoma(10, 11, 13, 39–42). We embarked on an exhaustive and all-encompassing analysis of 33 different global cancer types to discern the molecular characteristics of SMAD3. This analysis involved the utilization of diverse databases, including TCGA, GTEx, UALCAN, TIMER2.0, GSCA, and cBioportal. Our investigation aimed to elucidate SMAD3's role in tumor development and potential regulatory mechanisms, encompassing gene expression, prognosis, gene mutations, immunological infiltration, single-cell analysis, immune checkpoints, DNA methylation, RNA methylation, tumor mutation burden (TMB), microsatellite instability (MSI), tumor characteristics, and drug sensitivity.

Previous research has demonstrated that SMAD3 has a role in the progression and development of tumors and is expressed differently in cancer (43). Our findings demonstrate that SMAD3 is overexpressed in pancreatic cancer (PAAD) and adrenocortical cancer (ACC) compared to healthy tissues, consistent with prior investigations. Additionally, SMAD3 mRNA exhibited up-regulation in 13 other cancer types, with a down-regulation observed solely in 7 tumor types(BRCA, KIRC, OV, PRAD, SKCM, UCEC, and UCS). However, the data for SMAD3 expression differences in BLCA, CESC, KIRP, and READ were not statistically significant due to insufficient normal samples. The absence of normal samples also hindered the analysis of malignancies including DLBC, THYM, STES, UVM, PCPG, and MESO. Given the expanding dataset, this area warrants further investigation. Furthermore, our study revealed variable SMAD3 expression at different stages in PAAD, KIRC, BLCA, OV, KIPAN, BRCA, LUAD, and TGCT. We utilized immunohistochemistry (IHC) to validate the presence of SMAD3 protein in both cancer and normal tissue samples from PAAD, LIHC, GBM, and LUAD. In comparison to normal tissues, PAAD, LIHC, GBM, and LUAD exhibited higher levels of SMAD3 protein expression. Collectively, these findings suggest that SMAD3 may have a role in a variety of tumors and could serve as a prognostic predictor in pan-cancer.

Our Kaplan-Meier survival analysis, utilizing TCGA data, revealed that high SMAD3 expression was associated with poor prognosis in several cancer types, including LAML, PAAD, ACC, UVM, and LUAD. This finding aligns with earlier investigations, which have likewise reported a congruent predictive function for elevated SMAD3 expression in pancreatic ductal carcinoma and acute myeloid leukemia (44, 45). Conversely, elevated SMAD3 expression is linked to a favorable prognosis in patients with GBMLGG, KIRC, and KIPAN. Moreover, our study marked a pioneering investigation into the connection between SMAD3 promoter methylation and cancer. We uncovered a correlation between SMAD3 expression and DNA methylation, suggesting the potential utility of SMAD3 methylation level as a prognostic biomarker in cancer patients.

A genetic component contributing to a poor clinical outcome in LUAD and oral cancer is SMAD3 mutations (39, 46). We may conclude that SMAD3 is mutated in the majority of tumor types from the data analyzed in the cBioPortal platform. COAD, UCEC, and SKCM were the tumors with the highest incidence of SMAD3 variation, predominately mutation, followed byamplification. Moreover, SMAD3 expression correlates with CNV in GBMLGG, LGG, CESC, COAD, COADREAD, BRCA, STES, SARC, STAD, HNSC, LIHC, and SKCM. hence, we should focus on the association between SMAD3 gene mutation and these cancers.

An integral component of the tumor microenvironment (TME), tumor immunity is closely related to tumor immunity and drives tumor growth(47–49). The majority of TIME's components are immune cells(50, 51). Data extracted from the TIMER2.0 database unveiled a significant relationship between diverse immune cell infiltration and SMAD3 expression across various cancer types. Moreover, our analysis encompassed the assessment of immune and stromal components within the tumor microenvironment (TME), utilizing the immunological score and stromal score. Additionally, the pivotal parameter characterizing the status of the tumor microenvironment, specifically tumor purity, was indirectly assessed via the ESTIMATE score(52).

The findings indicate a robust positive association among the three immune infiltration scores in cancer types including LGG, COAD, COADREAD, PRAD, NB, READ, and UVM. In contrast, cancer types such as STES, KIRP, KIPAN, HNSC, KIRC, LUSC, LIHC, WT, BLCA, THCA, and ACC exhibited simultaneous negative associations. Furthermore, the study revealed a significantly positive association between immunomodulatory genes and SMAD3 expression in malignant tumors. These findings suggest that SMAD3 may contribute to the regulation of tumor immunity through its impact on immune infiltration.

Tumor mutation burden (TMB), microsatellite instability (MSI), and programmed death-ligand 1 (PD-L1) are frequently employed as markers to anticipate the efficacy of immune checkpoint inhibitors (ICIs). TMB, in particular, exhibits a strong correlation with the efficacy of PD-1/PD-L1 inhibitors, and the majority of clinical studies implementing TMB as a predictive marker have achieved their endpoints with remarkable success rates (53).

As a result, specific cancer patients may utilize the TMB marker to partially predict the efficacy of immunotherapy. Elevated TMB can also signify increased susceptibility of tumors to immune checkpoint inhibitors (ICIs). Notably, high SMAD3 expression demonstrated a positive correlation with enhanced TMB in GBMLGG, BRCA, LUAD, KIPAN, PRAD, KIRC, THCA, SKCM, BLCA, and DLBC, while it was associated with reduced TMB in SARC, HNSC, ACC, THYM, and PAAD. Moreover, microsatellite instability (MSI) pertains to the emergence of novel microsatellite alleles at a microsatellite locus within a tumor, in contrast to its normal tissue counterpart. MSI occurs due to insertions or deletions of duplicate units and is primarily attributable to functional flaws in DNA mismatch repair within tumor tissue (54). Mismatch repair (MMR) deficiencies can render tumors incapable of rectifying errors in DNA replication, leading to the accumulation of mutations and heightened microsatellite instability. The effectiveness of immune checkpoint inhibitors (ICIs) is directly proportional to the number of neoantigens recognized by immune cells. Furthermore, a positive correlation between SMAD3 and microsatellite instability (MSI) was observed in GBMLGG, COAD, SARC, SARC KIPAN, LUSC, and UVM. Additionally, we investigated the relationship between SMAD3 and tumor stemness. Cancer stem cells possess the unique ability to regenerate and generate a variety of tumor cells, making them indispensable in the contexts of tumor survival, growth, metastasis, and recurrence(55, 56). RNAss represents the stemness indices determined from expression data, whereas DNAss serves as the dryness index derived from methylation data. As the stemness index approaches 1, the degree of cell differentiation tends to decrease, while the characteristics of stem cells become more pronounced. Our analysis of SMAD3 expression and the tumor stemness markers DNAss and RNAss revealed significant negative correlations between SMAD3 expression and DNAss in four tumor types: COAD, COADREAD, BRCA, and UCEC. Conversely, In other tumor categories, there were evident and statistically significant positive correlations between SMAD3 expression and DNAss, including CESC, STES, THYM, THCA, UVM, and HNSC. In all tumors except PCPG, SMAD3 showed a significant inverse correlation with RNAss. These findings suggest that high tumor cell stemness is associated with low SMAD3 expression, which may promote tumor growth and metastasis. In the context of immune checkpoints, we identified over 30 immune checkpoints that exhibited positive correlations with most cancers, with the exception of KICH, ESCA, ALL, CHOL, and MESO. Additionally, our analysis indicated that low expression of SMAD3 often predicted a favorable response to immunotherapy. Collectively, these results demonstrate the close relationship between SMAD3, TMB, MSI, MMR, DNAss, RNAss, and immune checkpoints, highlighting the potential of SMAD3 as a novel biomarker for predicting the efficacy of immune checkpoint inhibitors (ICIs).

Although the contribution of SMAD3 to tumorigenesis is well-established, the precise underlying mechanism remains elusive. To elucidate SMAD3's biological functions in pancreatic adenocarcinoma (PAAD) and adrenocortical carcinoma (ACC), we conducted functional enrichment analysis. Our findings demonstrated that high SMAD3 expression was primarily associated with Interferon Gamma Response, Glycine, Serine, and Threonine Metabolism, as well as five signaling pathways, including the Notch Signaling Pathway, TGF-β Signaling Pathway, Wnt Signaling Pathway, Neuroactive Ligand Receptor Interaction, and P53 Pathway. The findings proffer an intriguing indication of the plausible implication of SMAD3 in the instigation of tumorigenesis and the intricate process of metastatic dissemination. Importantly, antecedent investigations into the realm of COVID-19 viral affliction have underscored the pivotal participation of SMAD3 in the pathogenesis of this malady, accentuating the conceivable therapeutic potency inherent in the abrogation or inhibition of SMAD3 as a prospective stratagem for disease amelioration (57). The prospect of SMAD3 serving as a focal point for anticancer interventions remains contingent upon a need for extended empirical inquiry. Furthermore, our investigation has unveiled potential therapeutic agents by scrutinizing the congruence between SMAD3 expression and drug sensitivity, thereby affirming SMAD3's standing as a viable candidate for tumor-directed therapies. Furthermore, a series of in vitro experiments conducted with hepatocellular carcinoma (LIHC) cells has illuminated the role of SMAD3 as an oncogenic entity, vigorously fostering the proliferation, migratory behavior, and invasive tendencies characteristic of LIHC cells. These findings, in a compelling manner, underscore the promotional influence of SMAD3 in orchestrating the proliferation and migratory capacities inherent to LIHC cells.

Several limitations are inherent to our study. First, there is a need for additional experimental confirmation, including immunohistochemistry and immunocytochemistry, to validate our findings. Second, the molecular regulatory mechanisms related to SMAD3 in various cancers require further exploration. Finally, it's essential to acknowledge the potential for systematic bias since we retrieved data from multiple databases for our analysis. These limitations underscore the need for more extensive research to fully elucidate SMAD3's role in cancer and its potential as a target for anticancer therapies.

Our study systematically examined the significant correlation between SMAD3 expression and clinical features, prognosis, mutation status, DNA methylation, RNA methylation, immune checkpoints, immunomodulatory genes, TMB, MSI, DNAss, RNAss, and drug susceptibility in several cancers, which helps us better understand the potential role of SMAD3 in pan cancer.

DATA AVAILABILITY STATEMENT

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary Material

AUTHOR CONTRIBUTIONS

ZC and WR contributed to the conception and design of the study. ZT, ZDD, JJJ and XJY drafted the manuscript. ZT and ZDD collected and analyzed the data. ZC and WR revised the manuscript. All authors contributed to the manuscript revision and read and approved the submitted version.

FUNDING

This study was supported by the National Natural Science Foundation of China [82060049 and 82170289] and Jiangxi Natural Science Foundation [20202BAB216009].

Author Contribution

J. Ferlay, I. Soerjomataram, R. Dikshit, S. Eser, C. Mathers, M. Rebelo et al., Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. Int. J. Cancer. 136(5), E359–386 (2015)
Y. Zhang, Z. Zhang, The history and advances in cancer immunotherapy: understanding the characteristics of tumor-infiltrating immune cells and their therapeutic implications. Cell. Mol. Immunol. 17(8), 807–821 (2020)
J. Ferlay, M. Colombet, I. Soerjomataram, T. Dyba, G. Randi, M. Bettio et al., Cancer incidence and mortality patterns in Europe: Estimates for 40 countries and 25 major cancers in 2018. Eur. J. Cancer. 103, 356–387 (2018)
J. Zou, E. Wang, Cancer Biomarker Discovery for Precision Medicine: New Progress. Curr. Med. Chem. 26(42), 7655–7671 (2019)
K.A. Brown, J.A. Pietenpol, H.L. Moses, A tale of two proteins: differential roles and regulation of Smad2 and Smad3 in TGF-beta signaling. J. Cell. Biochem. 101(1), 9–33 (2007)
P.C.T. Tang, J.Y.F. Chung, V.W.W. Xue, J. Xiao, X.M. Meng, X.R. Huang et al., Smad3 Promotes Cancer-Associated Fibroblasts Generation via Macrophage-Myofibroblast Transition. Adv. Sci. (Weinh). 9(1), e2101235 (2022)
B. Yu, F. Luo, B. Sun, W. Liu, Q. Shi, S.Y. Cheng et al., KAT6A Acetylation of SMAD3 Regulates Myeloid-Derived Suppressor Cell Recruitment, Metastasis, and Immunotherapy in Triple-Negative Breast Cancer. Adv. Sci. (Weinh). 8(20), e2100014 (2021)
P.M.K. Tang, S. Zhou, X.M. Meng, Q.M. Wang, C.J. Li, G.Y. Lian et al., Smad3 promotes cancer progression by inhibiting E4BP4-mediated NK cell development. Nat. Commun. 8, 14677 (2017)
Q.M. Wang, P.M.K. Tang, G.Y. Lian, C. Li, J. Li, X.R. Huang et al., Enhanced Cancer Immunotherapy with Smad3-Silenced NK-92 Cells. Cancer Immunol. Res. 6(8), 965–977 (2018)
Q. Zhou, L.R. Han, Y.X. Zhou, Y. Li, MiR-195 Suppresses Cervical Cancer Migration and Invasion Through Targeting Smad3. Int. J. Gynecol. Cancer. 26(5), 817–824 (2016)
J. Pan, Z. Liu, Z. Yang, E. Liang, C. Fang, D. Zhang et al., Circ_0001686 Promotes Prostate Cancer Progression by Up-Regulating SMAD3/TGFBR2 via miR-411-5p. World J. Mens Health. 40(1), 149–161 (2022)
Y. Su, W. Feng, J. Shi, L. Chen, J. Huang, T. Lin, circRIP2 accelerates bladder cancer progression via miR-1305/Tgf-β2/smad3 pathway. Mol. Cancer. 19(1), 23 (2020)
J. Sun, X. Ma, Y. Ying, W. Wang, H. Shen, S. Wang et al., SMAD3 and FTO are involved in miR-5581-3p-mediated inhibition of cell migration and proliferation in bladder cancer. Cell. Death Discov. 8(1), 199 (2022)
L. Gao, X. Tang, Q. He, G. Sun, C. Wang, H. Qu, Exosome-transmitted circCOG2 promotes colorectal cancer progression via miR-1305/TGF-β2/SMAD3 pathway. Cell. Death Discov. 7(1), 281 (2021)
E. Tarasewicz, R. Hamdan, J. Straehla, A. Hardy, O. Nunez, S. Zelivianski et al., CDK4 inhibition and doxorubicin mediate breast cancer cell apoptosis through Smad3 and survivin. Cancer Biol. Ther. 15(10), 1301–1311 (2014)
J. Peng, Z. He, Y. Yuan, J. Xie, Y. Zhou, B. Guo et al., Docetaxel suppressed cell proliferation through Smad3/HIF-1α-mediated glycolysis in prostate cancer cells. Cell. Commun. Signal. 20(1), 194 (2022)
T. Li, J. Fan, B. Wang, N. Traugh, Q. Chen, J.S. Liu et al., TIMER: A Web Server for Comprehensive Analysis of Tumor-Infiltrating Immune Cells. Cancer Res. 77(21), e108–e110 (2017)
M.J. Goldman, B. Craft, M. Hastie, K. Repečka, F. McDade, A. Kamath et al., Visualizing and interpreting cancer genomics data via the Xena platform. Nat. Biotechnol. 38(6), 675–678 (2020)
M. Uhlen, C. Zhang, S. Lee, E. Sjöstedt, L. Fagerberg, G. Bidkhori et al., A pathology atlas of the human cancer transcriptome. Science. 357(6352), eaan2507 (2017)
M. Uhlén, L. Fagerberg, B.M. Hallström, C. Lindskog, P. Oksvold, A. Mardinoglu et al., Proteomics. Tissue-based map of the human proteome. Science. 347(6220), 1260419 (2015)
J. Liu, T. Lichtenberg, K.A. Hoadley, L.M. Poisson, A.J. Lazar, A.D. Cherniack et al., An Integrated TCGA Pan-Cancer Clinical Data Resource to Drive High-Quality Survival Outcome Analytics. Cell. 173(2), 400–416e11 (2018)
X. Liu, H. Wu, Z. Liu, An Integrative Human Pan-Cancer Analysis of Cyclin-Dependent Kinase 1 (CDK1). Cancers (Basel). 14(11), 2658 (2022)
Y. Pan, G. Liu, F. Zhou, B. Su, Y. Li, DNA methylation profiles in cancer diagnosis and therapeutics. Clin. Exp. Med. 18(1), 1–14 (2018)
B. Yang, J.Q. Wang, Y. Tan, R. Yuan, Z.S. Chen, C. Zou, RNA methylation and cancer treatment. Pharmacol. Res. 174, 105937 (2021)
J. Gao, B.A. Aksoy, U. Dogrusoz, G. Dresdner, B. Gross, S.O. Sumer et al., Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal. Sci. Signal. 6(269), pl1 (2013)
C.H. Mermel, S.E. Schumacher, B. Hill, M.L. Meyerson, R. Beroukhim, G. Getz, GISTIC2.0 facilitates sensitive and confident localization of the targets of focal somatic copy-number alteration in human cancers. Genome Biol. 12(4), R41 (2011)
K. Yoshihara, M. Shahmoradgoli, E. Martínez, R. Vegesna, H. Kim, W. Torres-Garcia et al., Inferring tumour purity and stromal and immune cell admixture from expression data. Nat. Commun. 4, 2612 (2013)
Z. Zeng, C.J. Wong, L. Yang, N. Ouardaoui, D. Li, W. Zhang et al., TISMO: syngeneic mouse tumor database to model tumor immunity and immunotherapy response. Nucleic Acids Res. 50(D1), D1391–D1397 (2022)
D. Sun, J. Wang, Y. Han, X. Dong, J. Ge, R. Zheng et al., TISCH: a comprehensive web resource enabling interactive single-cell transcriptome visualization of tumor microenvironment. Nucleic Acids Res. 49(D1), D1420–D1430 (2021)
H. Yuan, M. Yan, G. Zhang, W. Liu, C. Deng, G. Liao et al., CancerSEA: a cancer single-cell state atlas. Nucleic Acids Res. 47(D1), D900–D908 (2019)
Y. Zhou, B. Zhou, L. Pache, M. Chang, A.H. Khodabakhshi, O. Tanaseichuk et al., Metascape provides a biologist-oriented resource for the analysis of systems-level datasets. Nat. Commun. 10(1), 1523 (2019)
A. Subramanian, P. Tamayo, V.K. Mootha, S. Mukherjee, B.L. Ebert, M.A. Gillette et al., Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc. Natl. Acad. Sci. U S A 102(43), 15545–15550 (2005)
P.H. Dear, Copy-number variation: the end of the human genome? Trends Biotechnol. 27(8), 448–454 (2009)
S. Sui, X. An, C. Xu, Z. Li, Y. Hua, G. Huang et al., An immune cell infiltration-based immune score model predicts prognosis and chemotherapy effects in breast cancer. Theranostics. 10(26), 11938–11949 (2020)
F. Chen, L. Shen, Y. Wang, Y. Chen, X. Pan, H. Liang et al., Signatures of immune cell infiltration for predicting immune escape and immunotherapy in cervical cancer. Aging (Albany NY). 15(5), 1685–1698 (2023)
S. Bagchi, R. Yuan, E.G. Engleman, Immune Checkpoint Inhibitors for the Treatment of Cancer: Clinical Impact and Mechanisms of Response and Resistance. Annu. Rev. Pathol. 16, 223–249 (2021)
P. Chen, W.H. Hsu, J. Han, Y. Xia, R.A. DePinho, Cancer Stemness Meets Immunity: From Mechanism to Therapy. Cell. Rep. 34(1), 108597 (2021)
F.X. Schaub, V. Dhankani, A.C. Berger, M. Trivedi, A.B. Richardson, R. Shaw et al., Pan-cancer Alterations of the MYC Oncogene and Its Proximal Network across the Cancer Genome Atlas. Cell. Syst. 6(3), 282–300e2 (2018)
N.I. Fleming, R.N. Jorissen, D. Mouradov, M. Christie, A. Sakthianandeswaren, M. Palmieri et al., SMAD2, SMAD3 and SMAD4 mutations in colorectal cancer. Cancer Res. 73(2), 725–735 (2013)
Y.T. Huang, A.C. Cheng, H.C. Tang, G.C. Huang, L. Cai, T.H. Lin et al., USP7 facilitates SMAD3 autoregulation to repress cancer progression in p53-deficient lung cancer. Cell. Death Dis. 12(10), 880 (2021)
Y. Zhao, J. Ma, Y. Fan, Z. Wang, R. Tian, W. Ji et al., TGF-β transactivates EGFR and facilitates breast cancer migration and invasion through canonical Smad3 and ERK/Sp1 signaling pathways. Mol. Oncol. 12(3), 305–321 (2018)
Y. Li, P. Wang, D. Ye, X. Bai, X. Zeng, Q. Zhao et al., IGHG1 induces EMT in gastric cancer cells by regulating TGF-β/SMAD3 signaling pathway. J. Cancer. 12(12), 3458–3467 (2021)
C. Huang, F. Hu, D. Song, X. Sun, A. Liu, Q. Wu et al., EZH2-triggered methylation of SMAD3 promotes its activation and tumor metastasis. J. Clin. Invest. 132(5), e152394 (2022)
K. Yamazaki, Y. Masugi, K. Effendi, H. Tsujikawa, N. Hiraoka, M. Kitago et al., Upregulated SMAD3 promotes epithelial-mesenchymal transition and predicts poor prognosis in pancreatic ductal adenocarcinoma. Lab. Invest. 94(6), 683–691 (2014)
J. Zhang, L. Zhang, H. Cui, X. Zhang, G. Zhang, X. Yang et al., High expression levels of SMAD3 and SMAD7 at diagnosis predict poor prognosis in acute myeloid leukemia patients undergoing chemotherapy. Cancer Gene Ther. 26(5–6), 119–127 (2019)
V.P. Sivadas, N.A. George, J. Kattoor, S. Kannan, Novel mutations and expression alterations in SMAD3/TGFBR2 genes in oral carcinoma correlate with poor prognosis. Genes Chromosomes Cancer. 52(11), 1042–1052 (2013)
S.R. Ali, M. Jordan, P. Nagarajan, M. Amit, Nerve Density and Neuronal Biomarkers in Cancer. Cancers (Basel). 14(19), 4817 (2022)
X. Lei, Y. Lei, J.K. Li, W.X. Du, R.G. Li, J. Yang et al., Immune cells within the tumor microenvironment: Biological functions and roles in cancer immunotherapy. Cancer Lett. 470, 126–133 (2020)
K. Yang, Z. Wu, H. Zhang, N. Zhang, W. Wu, Z. Wang et al., Glioma targeted therapy: insight into future of molecular approaches. Mol. Cancer. 21(1), 39 (2022)
C. Zha, X. Meng, L. Li, S. Mi, D. Qian, Z. Li et al., Neutrophil extracellular traps mediate the crosstalk between glioma progression and the tumor microenvironment via the HMGB1/RAGE/IL-8 axis. Cancer Biol. Med. 17(1), 154–168 (2020)
Q. Chen, B. Han, X. Meng, C. Duan, C. Yang, Z. Wu et al., Immunogenomic analysis reveals LGALS1 contributes to the immune heterogeneity and immunosuppression in glioma. Int. J. Cancer. 145(2), 517–530 (2019)
Q. Ju, X. Li, H. Zhang, S. Yan, Y. Li, Y. Zhao, NFE2L2 Is a Potential Prognostic Biomarker and Is Correlated with Immune Infiltration in Brain Lower Grade Glioma: A Pan-Cancer Analysis. Oxid. Med. Cell. Longev. 2020, 3580719 (2020)
A. Lin, J. Zhang, P. Luo, Crosstalk Between the MSI Status and Tumor Microenvironment in Colorectal Cancer. Front. Immunol. 11, 2039 (2020)
van M.J.M. Velzen, S. Derks, van N.C.T. Grieken, N. Haj Mohammad, van H.W.M. Laarhoven, MSI as a predictive factor for treatment outcome of gastroesophageal adenocarcinoma. Cancer Treat. Rev. 86, 102024 (2020)
F. Wei, T. Zhang, S.C. Deng, J.C. Wei, P. Yang, Q. Wang et al., PD-L1 promotes colorectal cancer stem cell expansion by activating HMGA1-dependent signaling pathways. Cancer Lett. 450, 1–13 (2019)
P. Nallasamy, R.K. Nimmakayala, S. Karmakar, F. Leon, P. Seshacharyulu, I. Lakshmanan et al., Pancreatic Tumor Microenvironment Factor Promotes Cancer Stemness via SPP1-CD44 Axis. Gastroenterology. 161(6), 1998–2013e7 (2021)
W. Wang, J. Chen, D. Hu, P. Pan, L. Liang, W. Wu et al., SARS-CoV-2 N Protein Induces Acute Kidney Injury via Smad3-Dependent G1 Cell Cycle Arrest Mechanism. Adv. Sci. (Weinh). 9(3), e2103248 (2022)

No competing interests reported.

figure1S.jpg
Supplementary FIGURE 1S | (A) Forest plots showed the correlation between SMAD3 expression and patient’s Disease-specific survival (DSS) in different cancers. (B) Kaplan-Meier curves for patient’s DSS stratified by different expression levels of SMAD3 in ACC, PCPG, LUAD and PAAD.
figure2S.jpg
Supplementary FIGURE 2S | (A) Forest plots showed the correlation between SMAD3 expression and patient’s Disease-specific survival (PFI) in different cancers. (B) Kaplan-Meier curves for patient’s DFI stratified by different expression levels of SMAD3 in ACC, PAAD, STAD, UVM, LIHC and STES.
figure3S.jpg
Supplementary FIGURE 3S | (D) Kaplan-Meier curves for patient’s DFI stratified by different levels of SMAD3 expression with DNA Methylation in CHOL and ACC. (E) Kaplan-Meier curves for patient’s DSS stratified by different levels of SMAD3 expression with DNA Methylation in UVM, SKCM, CHOL, LGG and ACC. (F) Kaplan-Meier curves for patient’s OS stratified by different levels of SMAD3 expression with DNA Methylation in SKCM, ACC, PAAD, LGG, LAML and UVM. (G) Kaplan-Meier curves for patient’s PFS stratified by different levels of SMAD3 expression with DNA Methylation in LGG, ACC, PAAD, UVM and CHOL.
figure4S.jpg
Supplementary FIGURE 4S、5S、6S | Correlation between SMAD3 expression and interstitial SCORE , ESTIMATE and IMMUNEscore.
figure5S.jpg
Supplementary FIGURE 4S、5S、6S | Correlation between SMAD3 expression and interstitial SCORE , ESTIMATE and IMMUNEscore.
figure6S.jpg
Supplementary FIGURE 4S、5S、6S | Correlation between SMAD3 expression and interstitial SCORE , ESTIMATE and IMMUNEscore.
figure7S.jpg
Supplementary FIGURE 7S | Heat Map of SMAD3 Association with immune regulator.

Download PDF

Version 1

posted

You are reading this latest preprint version

Multiomic characterization, immunological and prognostic potential of SMAD3 in pan-cancer and validation in LIHC

Status:

Version 1

Abstract

Figures

INTRODUCTION

MATERIALS AND METHODS

Gene Expression Analysis of SMAD3 in Pan-Cancer

Prognostic Analysis of SMAD3

DNA methylation and RNA modification analysis

Gene Mutation Analysis of SMAD3

Immune-Related Characteristics of SMAD3

Single-Cell Functional Analysis of SMAD3

Co-Expressed Genes and Enrichment Analysis of SMAD3

GSEA Enrichment Analysis and Drug Sensitivity Prediction of SMAD3

Cell culture and transfection

Western blot analysis

Wound Healing and Transwell Experiments

EdU assay

Statistical Analysis

RESULTS

Analysis of Gene Expression

Prognostic Value of SMAD3

Correlation of SMAD3 Expression with DNA Methylation AND RNA modification

SMAD3 mutation in pan-cancer

Analysis of Immune Cell Infiltration and Immunoinfiltration of SMAD3

Relationship between SMAD3 and immune checkpoints, immune modulators

Relationship between SMAD3 expression and tumor stemness, and genomic heterogeneity

Single-cell Analysis of SMAD3

Enrichment of SMAD3-Related Partners

GSEA Enrichment of SMAD3 and Relationship between SMAD3 expression and drug sensitivity

SMAD3 promotes the proliferation and Migration of liver cancer cells

DISCUSSION

CONCLUSION

Declarations

Author Contribution

References

Additional Declarations

Supplementary Files

Status:

Version 1